{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XSXNQ4JP66AUTGM376LMI4AVK7","short_pith_number":"pith:XSXNQ4JP","schema_version":"1.0","canonical_sha256":"bcaed8712ff78149999bff96c4701557d3d42af0bcf2105fd00eca50e90193be","source":{"kind":"arxiv","id":"2502.13145","version":2},"attestation_state":"computed","paper":{"title":"Multimodal Mamba: Decoder-only Multimodal State Space Model via Quadratic to Linear Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Haoran Yin, Hongyuan Tao, Qian Zhang, Tianheng Cheng, Wenyu Liu, Xinggang Wang, Yingyue Li","submitted_at":"2025-02-18T18:59:57Z","abstract_excerpt":"Recent Multimodal Large Language Models (MLLMs) have achieved remarkable performance but face deployment challenges due to their quadratic computational complexity, growing Key-Value cache requirements, and reliance on separate vision encoders. We propose mmMamba, a framework for developing linear-complexity native multimodal state space models through progressive distillation from existing MLLMs using moderate academic computational resources. Our approach enables the direct conversion of trained decoder-only MLLMs to linear-complexity architectures without requiring pre-trained RNN-based LLM"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2502.13145","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-18T18:59:57Z","cross_cats_sorted":[],"title_canon_sha256":"52dff5f5f9536fecfc6cbd90ac6a6fea5d9157d9d4798e18b9c600b341880db1","abstract_canon_sha256":"924a07e896d36c03176a092e7a0842980df462983f61d6e19f8363d213b2d229"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:33:45.230580Z","signature_b64":"bpwTg1ERpge1c2KMyM4iuR9hS5Mgy/89yQRWNhnQz8iqJCKvEvdfb80EU8KBEyWzdlSput73ZA2fN/jhUfmCCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bcaed8712ff78149999bff96c4701557d3d42af0bcf2105fd00eca50e90193be","last_reissued_at":"2026-07-05T10:33:45.229558Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:33:45.229558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Mamba: Decoder-only Multimodal State Space Model via Quadratic to Linear Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bencheng Liao, Haoran Yin, Hongyuan Tao, Qian Zhang, Tianheng Cheng, Wenyu Liu, Xinggang Wang, Yingyue Li","submitted_at":"2025-02-18T18:59:57Z","abstract_excerpt":"Recent Multimodal Large Language Models (MLLMs) have achieved remarkable performance but face deployment challenges due to their quadratic computational complexity, growing Key-Value cache requirements, and reliance on separate vision encoders. We propose mmMamba, a framework for developing linear-complexity native multimodal state space models through progressive distillation from existing MLLMs using moderate academic computational resources. Our approach enables the direct conversion of trained decoder-only MLLMs to linear-complexity architectures without requiring pre-trained RNN-based LLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13145","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2502.13145/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2502.13145","created_at":"2026-07-05T10:33:45.229700+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.13145v2","created_at":"2026-07-05T10:33:45.229700+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13145","created_at":"2026-07-05T10:33:45.229700+00:00"},{"alias_kind":"pith_short_12","alias_value":"XSXNQ4JP66AU","created_at":"2026-07-05T10:33:45.229700+00:00"},{"alias_kind":"pith_short_16","alias_value":"XSXNQ4JP66AUTGM3","created_at":"2026-07-05T10:33:45.229700+00:00"},{"alias_kind":"pith_short_8","alias_value":"XSXNQ4JP","created_at":"2026-07-05T10:33:45.229700+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00390","citing_title":"Zamba2-VL Technical Report","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17093","citing_title":"HEED: Density-Weighted Residual Alignment for Hybrid Vision-Language Model Distillation","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7","json":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7.json","graph_json":"https://pith.science/api/pith-number/XSXNQ4JP66AUTGM376LMI4AVK7/graph.json","events_json":"https://pith.science/api/pith-number/XSXNQ4JP66AUTGM376LMI4AVK7/events.json","paper":"https://pith.science/paper/XSXNQ4JP"},"agent_actions":{"view_html":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7","download_json":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7.json","view_paper":"https://pith.science/paper/XSXNQ4JP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.13145&json=true","fetch_graph":"https://pith.science/api/pith-number/XSXNQ4JP66AUTGM376LMI4AVK7/graph.json","fetch_events":"https://pith.science/api/pith-number/XSXNQ4JP66AUTGM376LMI4AVK7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7/action/storage_attestation","attest_author":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7/action/author_attestation","sign_citation":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7/action/citation_signature","submit_replication":"https://pith.science/pith/XSXNQ4JP66AUTGM376LMI4AVK7/action/replication_record"}},"created_at":"2026-07-05T10:33:45.229700+00:00","updated_at":"2026-07-05T10:33:45.229700+00:00"}